Table 2 Floral traits associated with the quantitative distribution of links in the static network and its sub-networks

From: Disentangling the role of floral sensory stimuli in pollination networks

(Sub-)network

AICsumnull

AICsumbest

Predictor

Deviance

P-value

Total

5166

4843

Trichromatic colour (rtri × θtri)

343.60

0.027

   

%Sesquiterpenes

352.80

0.038

   

Phenology (start × duration)

343.60

0.030

   

Surface apparency

551.10

0.004

Coleoptera

1188

1015

Phenology (start × duration)

79.94

0.009

   

Symmetry

90.78

0.003

   

Surface apparency

158.75

0.001

Diptera

1054

1021

Tetrachromatic colour (rtet  ×  θtet  ×  φtet)

76.78

0.069

   

Phenology (start × duration)

88.11

0.024

   

Apparency of total emissions

90.09

0.025

Hymenoptera—Bees

2035

1897

Phenology (start × duration)

103.80

0.143

   

Sesquiterpene apparency

163.60

0.011

   

Height apparency

192.70

0.005

   

Surface apparency

206.00

0.055

Hymenoptera—Wasps

353

310

Trichromatic colour (rtri  ×  θtri)

25.87

0.128

   

Surface apparency

63.17

0.012

   

Phenology (start × duration)

68.16

0.001

Lepidoptera

462

374

Surface apparency

63.09

0.005

   

Phenology (start × duration)

93.36

0.001

  1. Each subnetwork includes one of the major pollinator groups in the study community. The best multivariate-response generalized linear models (family: negative binomial) according to AIC are presented. The AIC value of the null model (~1) is provided for comparison. Significant P-values (>0.050) are indicated in bold. % Sesquiterpenes: the percentage of sesquiterpene compounds in the total floral emissions. For details on the predictors, see Methods, and the Supplementary Information
Back to article page